Synthetic Pricing Research

What It Is

Synthetic pricing research runs WTP and packaging studies (Van Westendorp, Gabor-Granger, conjoint) against synthetic-consumer-panels instead of real respondents, to rank many price/package variants before committing real, expensive, slow studies to the few survivors. It applies the SSR machinery to the highest-leverage, least-tested lever in SaaS.

The Two Non-Negotiables

Carried over from the SSR ingest, and especially binding here because pricing is high-stakes:
- Comparative use only — synthetic panels predict relative preference (package B beats A; price tolerance is higher in this segment), not absolute willingness-to-pay in dollars. Use them to choose what to field, never to set the price.
- Mandatory calibration-loop — pricing is a high-drift domain (domain-transfer-risk); every synthetic verdict that later gets a real outcome must feed back to measure agreement. No calibration record → advisory only.

The Workflow

SSR ranks the variant set → field the top 2–3 to a small real panel → fit the willingness-to-pay curves → decide with the small-N tools (Bayesian decision under MDE limits). Synthetic narrows the field; real buyers confirm; calibration keeps the synthetic stage honest.

How It Applies to Marketing Factory

This is a high-value new application of the factory's existing synthetic-panel + calibration capability: pricing studies are exactly the slow, small-N research that synthetic pre-screening accelerates, and the comparative-only constraint maps cleanly onto choosing which prices to test rather than setting them. It turns pricing from an occasional, intuition-led exercise into a recurring, agent-runnable study with a human at the final price decision.

Referenced from: pricing-packaging-and-wtp